AI Use: Claude Users Reveal Their Hopes and Fears

Is artificial intelligence contributing to people learning more and better? Or is it paving the way for global stupidity? User opinions vary widely. AI developer Anthropic has now presented the results of a survey in which 80,508 users of the language model Claude were asked about their hopes, fears, and experiences with AI. According to Anthropic, the interviews conducted in December 2025 via AI with participants from 159 countries and 70 languages are the largest and most multilingual qualitative study of its kind.

Because they are exclusively AI users, the Anthropic survey naturally carries a certain bias; this is also evident in the fact that, overall, more advantages than disadvantages tend to be seen. Viewed positively, however, respondents at least know what they are talking about through AI use.

Concern about hallucinations

What's the biggest concern? Unsurprisingly, the fear of unreliability and hallucinated results is the biggest risk for the reliability of AI applications, with 26.7 percent, followed by concerns about job loss and the impact on the economy (22.3 percent). 21.9 percent of respondents worry about a loss of human autonomy.

On the positive side, there is the joy that AI takes over routine tasks at work. While some studies find hardly any measurable productivity gains from AI in companies, many users hope for relief to concentrate on more meaningful work (18.8 percent), help with personal development (13.7 percent), or support in life organization (13.5 percent). In specific areas such as finance, this support is appreciated, but the fear of data leaks in AI-supported financial transactions remains a dominant issue, according to surveys.

However, the hopes and fears also often address what AI does to people. 33 percent of respondents see AI as a good learning aid. For example, a user from India raves about how he was able to successfully pick up math learning content again after his school days that he had never understood as a student. This boosted his self-confidence. But many also worry that they will lose skills in the long run because they let the AI do certain tasks. “I don't think as much as I used to. I struggle to put the ideas I do have into words,” quotes a frequent AI user from the USA.

Regional differences are also striking: people from countries in South America, Africa, and parts of Asia showed significantly more optimism about AI than those from Europe and the USA. This is apparently closely linked to the respective prosperity. In developing countries, AI is seen more as a tool for advancement, whether in education or as an entrepreneur. Western users, on the other hand, tend to organize their lives more with AI in comparison, but also have greater concerns about jobs and the economy.

Is AI taking over the wrong tasks?

More interesting than the numbers are the embedded quotes from the surveys. A user from Germany complains that the AI should rather clean his windows and empty the dishwasher so that he can paint and write poems. “Right now it's exactly the other way around,” he addresses a point that many complain about who work in creative professions. An entrepreneur from Honduras, on the other hand, sings its praises: He can relax while the AI does the work and increases his prosperity. It is his shadow, but a very long one.

(mki)